Global warming will lead to more intense, frequent, and longer-lasting extreme weather events, that have been associated with increased mortality. However, significant gaps remain in our understanding of the impact of climate change on health, including its impact on morbidity, the joint effects of multiple weather factors, effects o extreme cold weather, and whether there are especially vulnerable subpopulations. Little is known about the effect of extreme weather on birth outcomes in spite of biological plausibility suggested by animal experiments and limited human studies. The proposed study will fill these gaps by examining whether exposures to various extreme weather conditions or their joint effects during pregnancy are associated with adverse birth outcomes, including selected birth defects, preterm birth, and fetal growth restriction. To assess acclimatization and potential modification by air pollution, we will examine geographic variation and interactive effects of air pollution on the associations tested. Moreover, the study will assess whether pregnant women with certain chronic diseases, those taking heat-sensitizing medications, those of low socioeconomic status, those who smoke or drink alcohol, or those with outdoor occupations are more vulnerable to extreme weather conditions. Finally, we will estimate weather-attributable risks for pregnant women and women with certain characteristics, project the birth outcome burden of climate change, and develop vulnerability maps. We will use the National Birth Defects Prevention Study (NBDPS), the largest US population-based case-control study of potential risk factors of birth defects. The birth defect cases include infants with confirmed, selected major birth defects excluding genetic causes (N=23,333) and the controls are non-malformed live-born infants, randomly selected from each center (N= 8,494 controls). Other pregnancy outcomes will be selected from within the NBDPS control group, such as preterm (< 37 weeks) including severe preterm (<32 weeks) and moderately preterm (32-37 weeks), and small-for-gestational-age (below 10% of birth weight for gestational age). Extreme weather indicators defined by the 90th percentile of average summer temperature or the 10th percentile of winter temperature will be evaluated by intensity, duration, frequency, and timing. Other indicators including heat waves and cold spells, the full spectrum of temperature, humidity, air pressure and wind as well as a composite weather index using the Spatial Synoptic Classification system will also be examined. A two-stage Bayesian hierarchical model will be used to first examine the association in each region and then to control for regional characteristics to obtain a nationwide estimate. K-means cluster analysis and Monte Carlo methods will be used to estimate weather factor clusters and uncertainty, respectively. The findings may identify a vulnerable population which has been ignored, and thereby guide climate intervention and adaptation. The multidisciplinary research team has a unique opportunity to use the data already collected and geo-coded through the NBDPS and other ongoing projects, which ensures the project is feasible and sustainable.